The Simulation Hypothesis is Pseudoscience 

This is... Exactly what the simulation would WANT us to believe.

I absolutely love your strict sense of epistemological truths by saying that illogical things are not necessarily wrong. We just can't verify them scientifically

Thanks Sabine - I did think that when Tyson brought out the 50-50 odds about simulation, it sounded like something being thrown in from nowhere rather than actual science. I expected better than that from an astrophysicist

Just discovered your channel, you remind me of so many of my favorite teachers from growing up, and it makes your videos and learning from them really cozy. Keep up the great work!

Hitchhiker's Guide to the Galaxy: "There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable."

I really appreciate how she just basically told us the conclusion of the video in the title. No misleading or baiting title.

The simulation argument concludes we are most likely NOT in a simulation. Sometimes it seems like no one has actually read the paper. (I really hope im not wrong on this one lol I've read the paper several times it's pretty interesting but maybe i forgot)

I appreciate that the hypothesis we live in a simulation is untestable, and thus, fundamentally unscientific. However, that doesn't make it impossible. A big shortcoming in the argument against the simulation hypothesis that's presented here is that it assumes that the higher level reality must abide by the same laws of physics and use the same mathematics that we have. There's no reason to assume this. For example, for all we know, in a higher level reality, information density and/or processing power/speed could be infinite, and maybe spacetime itself doesn't exist in any form that we'd even recognize. The laws of physics that we struggle to comprehend may just be some made up rules that a superintelligent being created in their spare time, with the "real" universe profoundly more complex than we can comprehend.

Yo whoever is controlling my character, there's no need to choose hardest level everytime.

I think smbc's take is funniest: if there are many arguments and most arguments are wrong, then the simulation hypothesis is likely to be wrong.

I've done the simulation hypothesis argument too many times. I have settled on a single anti-SH argument. A single argument which is to consider ....... a single droplet of water. If you simulate a droplet of water you can do an excellent job with several graphic tools and AI models. BUT if you take the simulated droplet, extract the profile with high resolution, determine the first and second derivatives along the profile and then try to verify the Laplace Young equation it will be off by ... a lot (at least a couple of orders of magnitude). Point being. It is EASY to SIMULATE a physical object to the point that is seems real to casual observation but it is extremely HARD, borderline impossible (due to things like resolution limits and computational round-off error) to simulate the same physical objects to the point that they seem real under determined physical analysis. So, we are NOT a simulation, and to convince yourself of this take a photo of an actual droplet and try to extract the LY equation. That exercise will be a humbling experience.

Wow, thanks for the bonus information/ explanation of climate modelling/prediction. The 10 km resolution description you gave helped me to have a better understand, that climate modelling systems use low res approximations, that are linked with regular observation to prove whether or not the low res approximation is accurate enough and useful.

The main concern I have about the simulation hypothesis, other than being unfalsifiable, is its recursive nature. If we would indeed be living inside a simulation, there's no reason whatever is running the simulation isn't inside a simulation itself. It's turtles all the way up.

My understanding of the "simulation hypothesis" was that it was a thought experiment and much more philosophical than scientific.

The problem I have with Bostrom's formulation is the same problem I have with the Creationist Fine Tuning argument. The fact of the matter is, we're here, right here, right now. That's the starting point. The probability of the circumstances right now is 100%. You don't get to propose a hypothetical "what if" version of the past and assume is is a roll of random dice to get from there to here, and then form a probability argument. If you want to assert the distribution of possibilities is random (instead of selective, biased, or causal), then you should have to prove it instead of just get it as a free assumption in your argument. Creationists are not allowed to assume that constants "could have" taken any possible values, and then say the probability of their current values is extremely small. The fact is, the values are what they are, and if you want to assert they "could have" been different, then that is up to you to prove. So when it comes to Bostrom's argument, where does he get the free pass to propose a totally fictional and abstract space of conscious civilizations, and then get to toss us in that abstract space like a plinko lottery instead of starting with our actual circumstances as the ground truth? So just like the creationist, if Bostrom wants to say there "could have" been these other civilizations that perform simulations, that is up to him to prove, not just get a free assumption as a possibility. We're not some abstract civilization in a 3 step armchair fart, we are a highly specific set of details and circumstances and context, and *every* single variable should factor into a continuously updated Bayesian inference. The probability of you as an individual getting cancer is not the same as the probability of the average person getting cancer. Every single individual risk factor plays a part as a variable, with your non-statistical individual chain of causal circumstances being the ultimate determination of the outcome. The fact of the matter is, Bostrom's argument when applied to our actual reality and evidence we have today says there is zero chance of simulation. If the civilization performing the simulation must be "outside" our physical universe in some manner, and therefore excludes all physical civilizations on Earth, then the number of civilizations we have evidence for on step 1 is exactly zero, and (even when you include our own civilizations) the number of simulators for step 2 is exactly zero, therefore the probability of us being simulated is zero.

I really like this youtuber's delivery and style, deals with complex subjects intelligently but with great clarity.

Her argument that it's not easy to reproduce the foundations of physics with code is valid only if we accept that the same foundations are true outside the simulation too. But if we live in a simulation the physics of our universe may very well be only an artifact of the simulation

If we are living in a simulation, then the laws of physics as we know them are really just the rules of the simulation. Therefore, we wouldn't actually know the real laws of physics in the "real" world, so how can we say it wouldn't be possible to simulate? In the end though, it really doesn't matter. Even if we are in a simulation, that has exactly zero implications for how you live your life. You can't get out of the simulation, and you can't break the rules of the simulation (i.e. break the laws of physics). If the designers decide one day to just shut it all down, not only is there nothing you could do about it, you also wouldn't even be aware it had happened. You would simply cease to exist. This is what happens to everyone eventually, anyway. Whether by the simulation ending or you dying, you aren't going to be aware that it's over. What I'm getting at is that this reality or simulation we live in is what we have, and it makes no difference which it is. So, my take on the simulation theory is that I don't really care, because it doesn't matter either way.

Here's an idea to consider. 1- I can run a virtual machine (VM) on Windows 7, which simulates a physical machine with software, albeit slower due to *power and speed limitations* 2- Inside the Windows 7 VM, I can run Minecraft, just as I would on the host machine, albeit slower. *There is no difference in logic* between the VM and host machine. 3- I can build *computers within the Minecraft world* , but I cannot create a *computer that can run an exact copy of Minecraft within Minecraft* . The *in-game logic and physics would prevent this* , even if it were possible, it wouldn't be a perfect one-to-one copy. Where I'm going with this is, just as the in-game physics of Minecraft prevent us from building a computer that can simulate the Minecraft world, we can expect that the physics of our universe may also prevent us from simulating it. The physics of the "outer world" may be more permissive, allowing its "inhabitants" to create our universe as a computer game for their children to play with. As of now, we cannot prove or disprove this idea, which indeed places it within the realm of pseudoscience. However, to me, this idea makes more sense than the notion of a god simulating his own universe one-to-one. Unless I'm wrong, in which case I would be interested in learning how.

When I finally read the details of the simulation hypothesis it seemed clear to me that it's based on a lot of unfounded assumptions about the limits (or rather, lack thereof) of computation.

"I had a dream: I was a butterfly. But now I have a problem: did I dream about being a butterfly, or am I the dream of that butterfly?" [Chuang Tzu]

"It's just normal paranoia, everyone in the universe has that" Douglas Adams

6:55 actually such a concept exists and it's called LOD (Level Of Detail) in game dev. LOD main goal is to improve the performance BTW superposition state till the wave function collapse could be a result of such a system 🤷‍♂️

Mario concludes that he cannot be in a simulation because he is unable to build an NES out of mushrooms and coins

This channel should be called, "um actually"

Keep in mind. In the end it doesn't matter if our experience is real or simulated; It's real to us, and affects us just the same.

The simulation doesn't need to produce the whole universe, or even a small localised part of the universe. It doesn't need to maintain absolute consistency with known physical laws. All it needs to reproduce is the input of your senses and your thought processes about that input. Even the most advanced physicist only ever directly perceives a tiny fraction of all that they 'know' to be true. And the brain already has well-known circuits dedicated to making sense of inconsistent or confusing stimuli. It's pretty much the primary function of perceptual systems. Our brains synthesise and confabulate information with such rapidity and ease, and then engages other circuits specifically aimed at making us think those confabulations are true. In a sense, we are all living in a simulation because that's literally what our brain is doing for us in every waking moment.

As Sabine explains, the simulation idea of the universe is a bit silly. However, there is a hint of correct insight, in that conscious experience is a simulation (i.e. model) of the world by the brain that is being constantly updated by incoming sensory information, in a Bayesian updating sense. Specialised parts of the brain contribute parts of the overall conscious experience, e.g., the colour centre in the brain, the movement centre in the brain, etc. Damage to these areas changes our conscious experience. Our neuroconsciousness is a simulation of the world as it is happening, we are our simulation (e.g., red is part of our simulation, there is no red 'out there'), the brain is the mechanism of the simulation, strange experiences such as hallucinations in external space can be accounted for by dysfunctions in the brain's simulation - but there is no simulation underlying external physical reality. Also, time does not exist, its an ordering mechanism used by our brains to construct conscious experience, but that's a longer story.

I think the simulation hypothesis is one of those cases where we can quote Carl Sagan: "It pays to keep an open mind, but not so open your brains fall out."

I'm still waiting for one of the knights in the Age of Empires game to ignore my direction into battle, turn around and say, "How bout getting out of your chair and you going into battle."

You shared knowledge of interesting simulated algorithms and state of matter on both sides of probable acted out theories is well presented ... Thankyou

@6:48 Yeah most proponents of the simulation hypothesis I hear from often suggest the fact that if we can do back-face culling and other optimization in Video games, somehow that equals we can do that for a simulation of a universe without providing any concrete idea how to connect the dots (perhaps a better understand of Neurology and physics might answer that). Although I don't believe in the Simulation Hypothesis. As someone who is interesting in both computer science and Physics, the idea of the simulation hypothesis and the holographic universe are very interesting to me. Because those are one of the few time those fields have overlap. When I think of the simulation hypothesis it makes me think of the speed of light, the Planck length, and Planck time in our universe compared to the Delta T values we use to run our simulations of N body problems (like when we model galaxies colliding). The size of Delta T effectively limits the resolution of a simulation. The smaller Delta T the better, but then we run into the issue of the speed of our computers and potentially finite time in the universe for our computer to calculate. Does that mean when ever we choose a Delta T that give us a compromise between meaningful results and practical run times, are we effectively simulating a universe where either the speed of light is slower then ours and/or the Planck Length (and to extension the Planck time) are larger? More on the computer science end of the spectrum I wounder if you actually wanted to run an ancestor simulation would limiting your program to only simulating the internal perception of reality a conscious being experiences (the philosophical equivalent of a brain in a jar receiving specific electrical stimuli) instead of the whole "outside world", if that would make it any easier to compute or if it would still be the same amount of data?

You're assuming the simulator universe has the same laws as our simulated universe. But it's necessary that all simulated universes are simpler than the universe that is simulating them, so the fact we can't sinulate our universe inside our universe isn't a problem for the hypothesis. All simulated universes must be simpler than the universe running the simulation, or it wouldn't fit. It may be impossible to simulate our universe with our laws, but easier with whatever characteristics the universe simulating ours has Though if the universe that simulates our universe runs by different laws, then you don't know if the simulation hypothesis would be valid there, which does seem like a possible flaw.

I've always viewed the simulation hypothesis in the same category as Boltzmann Brains, Brane theory, and even string theory as "fun thought experiments that shouldn't be taken too seriously until there's any actual evidence"

Actually there are many videogames that only generate what the player is interacting with. Kind of like how a partical is a wave until observed/measured/path information is consciously available

As a programmer, and having a best friend that's also a programmer, we both came up with this idea as late-teens. It was fun to kick around, and to toy with, as it meant that somewhere in the code, we could turn on "god mode" and possibly "no-clip" modes. Mind you, we never really believed it, because a lot of observation just didn't line up, but it was really easy to draw "conclusions" that we were in a video game. Also, having played around with that, when I first heard about Simulation theory, I laughed and dismissed it as 20 years too late and fraught with holes.

"It gives me hope that things will be better on the next level", best summary of religion I've heard.

From my humble understanding of the simulation hypothesis (SH) I thought the "calculation of details" issue had been addressed. They say that the outcome does depend on whether someone is paying attention or not and point to the double slit experiment for this. In the double slit experiment we get an interference pattern, if no one is checking which slit every single photon went through. But, if we do observe for every photon which slit it went through, then the interference pattern vanishes. For proponents on the SH this points to the "physics engine" not caring about these details as long as no one is looking. PBS space time has a fascinating video about the "quantum eraser" where they show how this even seems to be working backwards in time. In a clever setup the photon hits the screen and only after that they measure which slit it went through. This is enough to make the interference pattern vanish as well. For the SH this means that it simplifies reality at first, but if it turns out later that the details were important, it goes back and puts them in. This might take some time, but as we are part of the simulation, we don't notice, because we are put on hold until this has been done. To be clear, I don't believe in the SH, but it is not easy to dismiss. My main gripe with it is that the SH is unfalsifiable and that puts it into the category of pseudoscience.

Very well done. I always thought the simulation theories were bogus us for two reasons. 1. That you would generally need a system bigger than the system you are simulating if you want to simulate it exactly and nested simulations would become impossible at some point. 2. There is the stopping problem especially for nested simulations. If any of the simulations above us stops we also cease to exist. The longevity of our universe seems to show this is not the case. Both these point to the real world more likely being base reality rather than some kind of simulation.

Simply & brilliantly presented!

0:55 The being outside the simulation doesn't have to be all knowing or omniscient. They just have to have access to more information and and more control over our universe than we do. For instance an outside developer could make changes by observing and allowing an event to play out giving them knowledge of an occurrence. They could then rewind the simulated event and make a small changes and do it over and over again till you get the perfect results. For the beings inside the simulation they would only perceive The single pass through but the beings manipulating the simulator reality will be viewed as omniscient and all powerful when in fact there anything but in their layer of reality.

“I dare you!” Lol you’re great and I love your channel.

I think you're missing the crux of Bostrom's Hypothesis, it was that you **cannot** out rule that we are living in a simulation, not that he proves we are.

I think it is likely that the universe arises out of infinitely large probabilities (relative to everything else) of certain sets existing that have properties that self promote their growth and stability. For example, having the ability to change (time), and allowing the set to have operations which allows duplication, interactions, etc. can increase the relative amount of sets of this type. Sort of an evolutionary + probabilistic model of power sets.

The simulation argument is a statistical argument with a sample size of 0.

Summary : "We are a simulation" has no evidence. "We might be a simulation" can't be proved impossible yet.

The argument against any kind of objective reality is a good one. That no matter what we think we are measuring or observing is still only taking place within our mind’s perception of those observations. Die Gedanken sind frei, und alles, was wir wirklich haben.

There's another major problem with the simulation hypothesis (please tell me if/why this critique isn't sound). It assumes there's infinite energy available in the "root reality" to run the nested sub-realities. Let's assume we're the root reality and imagine a "video game within a video game" scenario, similar to how people can simulate logic gates and therefore computers within Minecraft. Suppose someone uses this virtual Minecraft computer to run its own game, albeit a very slow-running one. Running the "game within a game" isn't free. It requires energy here in the root reality to run them. This would necessarily apply to every sub-reality, no matter how deeply nested, resulting in a cumulative total of energy needed to run them all. Unless the root reality (as well as every sub-reality) either has access to infinite energy or can run calculations without using any energy, this puts a physical limit on how many layers of reality can exist. Thus, the rationale used to conclude that we are almost certainly a sub-reality as a consequence of probability is flawed.

Here's a fun idea. The double slit experiment exposes a code optimisation that approximates light as waves until it is carefully observed. Video game developers use coarser approximations when possible to save computation all the time.

What we experience personally is a simulation in the truest sense.

Isaac Asimov - "The Last Question" is basically the Simulation Theory. He wrote it in 1956. It is a plausible explanation. Why? Heisenberg: "...the smallest units of matter are not physical objects in the ordinary sense; they are forms, ideas which can be expressed unambiguously only in mathematical language."

I guess you could argue in circles forever on this one, but just for the heck of it... If we are in a simulation, then everything we understand about our theories and observations are of a simulated, possibly imperfectly so, universe. What is possible to simulate from the perspective of the people living in the simulation may be irrelevant in the outside universe of the programmers. We are already on the cusp of having open world games with AI driven Non Player Characters. With enough variables in a personality matrix for each NPC, you could have a world filled with unique "people" with unique personalities... and then if AI eventually became conscious, You would have sentient people living in a simulation we created, asking these same questions... It's fun to ponder :)

who ever is running the simulations' operating system needs to update the anti-virus protection, given the last year or so.

5:08 Plot twist: our universe not just a simulation, it's also written in Javascript. 😱😱😱

There's always a level in science in which you're forced into belief; it may not be a healed blind man, but for instance speaking of energy pretending to know what energy is, speaking of force pretending to know what a force is, and so on.

Maybe the reason we don't have a unified theory is that it is impossible to compress the information of a universe in a simplified model. And to think you can do this recursive is insane. We don't have limitless computing power, so the hypothesis ends here.

"In science we require explanations for how something works." Do we? What is the explanation for collapse of the quantum wavefunction? How does entanglement work? Why is the strong force 137x larger than the electromagnetic force? What explains the ratio of the electron mass to that of a proton? What can we say about the interior of a black hole? Why can't we see the large majority of the matter and energy necessary to explain the large scale structure and dynamic behavior of the universe? There are huge areas of QM and GR where we don't have explanations, yet we find it useful to use them nonetheless. When legit physicists take ideas like multiverses, the 10^500 landscape and Boltzmann brains seriously, its rather slippery to draw neat lines of this type between science and pseudoscience. Probably the best argument against the simulation hypothesis is the one you brush past at the beginning: Neil Tyson might believe it.

Also because of the whole "not being able to falsify or verify" and "not being able to make predictions with it" things

No. WE approximate, because WE don't know the laws of our universe. The being which runs the simulation knows the rules and doesn't have to approximate. The simulation doesn't have to be "perfect" and doesn't have to allow the beings inside it to even be able to guess those laws. Also the simulation doesn't have to be complete - surely most of our computer simulations are approximate models which are failing in some conditions or outside of some domain. We either simplify the rules when it doesn't matter or we don't compute the "universe" when and where nobody important for us is affected by this "optimization". Still the being inside it wouldn't have any idea about it... And, yes. We need to assume all those things you mentioned - still it doesn't cross the "science/faith" border. The argument just gets more "IFs". Another set of P(x) probabilities to include in the multiplication, not changing the general idea. If you don't agree with it something tells me that you'd also call Copernicus telling about his heliocentric idea "an unscientific believer". He didn't know HOW it is possible to force planets to orbit the Sun. From his perspective circular heliocentric orbits with all those epicycles could look like unnecessary complication, without any scientific justification. The "faith" border is only reached when someone says some numbers: "over 50/50 chance", for example.

I think the thing that bothers me most about the Simulation Hypothesis is that it is inevitably solipsistic. When you do the math for exactly how many bits it would take to store the information in the Planck cubes of even a small portion of the universe, it's absurd. No computer memory is large enough even theoretically to pull it off. Let's not even get into the computing power it would take to compute one Planck second of time through the entire universe. Then you have the problem of points of view. The simulation is simulating all consciousnesses that exist, and all that have existed? Or just me? Well the only one that matters to me is mine, right? So it could just simulate my consciousness and then suddenly the entire universe revolves around me. One other thing I've noticed is that people who say they believe in this all have pretty narcissistic personalities. Ultimately I have to agree with Dr. H here, that there's no way this hypothesis can by scientific, because it really can not be disproven.

For the simulation hypothesis to work, a programmer doesn't need to simulate billions of people, they only need to simulate one. Just me. Me personally. I know all the rest of you are just characters.

to a man with a hammer, every problem looks like a nail. to a man with a computer, every problem looks like a simulation.

This video is really needed! Another point she could make is that Bostrom is assuming that a human brain can be massively compressed in silicon etc. If consciousness actually comes from microtubules or RNA, then the most efficient thing in the universe for simulating a human brain is a human brain. Sometimes there's no substituting for the real thing.

Interesting thing about your point about the resolution of the simulation. If it goes down with simulations being embedded into one another, does that mean it would increase as you expand outside of each simulation?

I don't think you should limit the simulation to being run on the type of machines we have developed so far. Obviously the 'computer' running the simulation of our universe is a tad more advanced.

Recently, Sabine says she has learned to love pseudoscience. Sabine has also shown that the simulation hypothesis is pseudoscience. Therefore, Sabine has learned to love the simulation hypothesis.

A simulation hypothesis can't be pseudoscience because it's just a hypothesis

I realize there is an a apriori argument refuting simulation hypothesis. It comes from a published paper. I think the paper points out that in line with the thinking in this video, the simulation hypothesis is not a scientific hypothesis; And if it can be refuted, it can only be refuted non-empirically. And I think it does give a compelling case why we are not in the simulation, by showing that the thinking itself that we are in the simulation is nonsensical or incoherent.

I love how Sabine says 'hypothesis.'

This episode is flawed from the base. Bostrom’s argument is not the one displayed and commented. It is more like this : 1- civilization(s) end before able to do simulations 2- civilisation(s) do not want to start a simulation 3- we are in a simulation Choose which one you want .

Interesting Video. I think that it's somewhat semantics on whether this is a "simulation" or a "real world" with real laws. One difference might be that presumably a sim would have some limits to its fidelity, and some magic to save computation and memory. But don't we see this? The planck scale is the distance limit. And the uncertainty principles define the max precision of the advanced machines running the Universe. The quantum fuzziness prior to observation is a great way to save computation and memory akin to how selective rendering is used in computer games. The discretization part could merely be a design choice though (we do this all the time); The fuzziness to me is more indicative of a sim. Another diff could be that a sim almost certainly implies a higher power being involved while those believing in a "real world" differ on whether there was a god or things merely arose from a big bang. However, those people can't tell us where the conditions that created the big bang came from. So not too different on this front really. I think we'll easily prove that it's possible we live in a sim. But to ever really know for sure might remain a matter of personal belief.

My biggest concern about the hypothesis is about definitions and clarifying what is meant by "simulation". It's not just semantics, as noted in this video. When we talk about a simulation, whether computer, scale model, or other types, what we mean is that we've created a limited and bounded model that we run through a process meant to simulate -- an analogy with similarly properties -- to something else in the " real" world. But these simulations still follow the laws of physics. A scale model of a ship in a wave pool is still a small boat riding real waves. Computer programs are still electricity running through circuits creating patterns. What makes the "simulations" is entirely within the mind of the operator as to how the information will be used. How people use the simulation hypothesis seems more broad in the sense of a description of the "laws" or our universe being constraints that are a subset of the physics of the higher layer of "reality". An obvious objection then is simply Occam's Razor. The simulation hypothesis requires " us" to have a physics nested within the physics of a higher level reality. And, the same arguments also mean the higher level reality is also likely a "simulation" in an even higher level reality. Ad infinitum. It's simulations all the way down. Adding one layer of reality above us adds complexity without any new explanatory value, and brings along with it all of the infinite layers above it for infinite complexity. It fails Occam's Razor badly, I think.

This lady makes learning things I know nothing about very entertaining for me.

The weather outside is frightful, but Sabine's so delightful...

Beautifully put Sabine Hossen welder. Everything falls into the yes and no categories. Everything is as Neutral as Natural, acting within the realms of the Infinite possibilities where the odds always end up at 010. ALL at 10nce, Timing the odds.. A New Paradigm

I'm a software developer and had this idea we could be a simulation, for many years already, I didn't know someone would say some day, believe it almost like a religion, that's crazy. A possibility, it's just that. And please stop trying to understand something like that from our technology point of view, quantum processing don't have to even be mentioned, it wouldn't exists outside our "reality" if we are a simulation, outside reality could be totally unexpected from us. Imagine an AI NPC of a virtual world without own access to internet in any how, trying to figure out how we the creators are, and then arguing that for them to be simulated, we (the creators) need to have an insane amount of chaos and soul stones at least level 12, so he concludes it's not possible.

I've never thought of it as a scientific argument. Just an argument on probability with very honest assumptions. Like how many theories are based on the universe being infinite?

Humans: make simulation to simulate reality also humans: "wait reality looks like a simulation"

About to hit 400k subs, great job ! 👍🏼

Here's Chatgpt's critism of this video The video makes an interesting argument about the simulation hypothesis and takes a rational and empirical approach to criticize it. Here are several areas that could be further examined for possible criticism: 1. **Assumption of Civilization Interest**: The argument Bostrom presents and the video accepts, presupposes that advanced civilizations would be interested in creating simulations of conscious beings. This is a cultural and value assumption that might not necessarily apply to alien civilizations. Their interests, technology, and ethics could be vastly different from ours, making this assumption potentially flawed. 2. **Unclear Definition of Simulation**: The video tends to imply a somewhat classical understanding of simulations and computers. We tend to think of simulations as human-made, computer-driven systems. However, what if the very nature of existence or consciousness involves some sort of inherent "simulation" process that is beyond our current understanding of simulations and computation? 3. **Consciousness in a Simulation**: The argument dismisses the complexity of consciousness and how it might be simulated. It says, "consciousness is simply a property of certain systems that process large amounts of information." This oversimplification disregards many philosophical and scientific debates about the nature of consciousness, its correlation with information processing, and if it could be artificially replicated in a simulation. 4. **Quantum Physics and Relativity**: The video asserts that we can't reproduce the laws of quantum physics and General Relativity with a computer simulation and thus claims this to be a weak point of the simulation hypothesis. While this is true at the moment (as of my knowledge cutoff in September 2021), it doesn't mean it will always be the case. Our understanding and computational abilities could improve to allow us to simulate these laws accurately in the future. 5. **Computational Power**: The video critiques the simulation hypothesis for not explaining how one could compute the immense detail of a simulated universe without inconsistencies. However, the video does not consider the possibility of the simulating civilization having access to technology and computational resources far beyond our own comprehension. 6. **Mixing Religion and Science**: The video concludes by suggesting that belief in the simulation hypothesis is akin to faith, implying a dichotomy between faith and logic. While the distinction between religion and science is often useful, it's also reductive, especially when dealing with theories and ideas at the edge of our understanding. It's possible to believe in the simulation hypothesis while also acknowledging it's currently unprovable and highly speculative. In conclusion, the video makes compelling points but perhaps does not fully consider the range of possibilities that could arise from civilizations or technologies far beyond our current comprehension. It also tends to oversimplify complex aspects such as the nature of consciousness and the potential abilities of advanced civilizations.

The thing is, if you made a simulation of a (part of a) universe, you don't have to follow any specific set of natural laws exactly. From within that universe, whatever laws were there, approximate or not, would just be "the laws" of nature. I agree the hypothesis does make the assumption that whatever is running the simulation may need to intervene. But the thing is we would be totally unaware if the simulation was paused, rewound, tweaked, and fixed in case any flaw was observed in universe that leads to the discovery of the simulation. I don't think that's as hard a problem to solve as Sabine does. If you're running the simulation of course you can detect when something with agency in the sim starts seeing evidence of the sim. I do think the idea that anyone will ever be able to run any sims on the equivalent of their laptop is ludicrous, these would be planet scale computers simulating tiny pieces of a universe to high fidelity. And you wouldn't be able to run a simulation of a universe where it was possible to do much interstellar travel, or you wouldn't run it that long, or you'd kill off any progress made to that. I guess my point is just that I don't think it's hard at all to argue that everything we observe *could be* simulated, it is all within the realm of possibility. But to take it on faith, well, that's just faith.

I always love how the people like Elon musk and Neil deGrasse Tyson throw around these probabilities like “1 in 1000000000” or “better than 50/50”. Like how did they get their?? What formula did they use to calculate this probability? Does it work on known facts or do they add unfounded assumptions to make it work? It’s, like, explain yourselves!

I love Sabine and normally agree with her on just about everything, but not this time. And yes, I get how that sounds, like I'm defending a transgressed religious principle. The difference is I have good reason to disagree. Sabine says the Simulation Hypothesis makes unreasonable assumptions and doesn't explain how a computer code for the universe would work. Well no, if we could do that, we would be running such simulations already. The hypothesis stipulates very-highly advanced civilizations with arbitrarily large amounts of both computing expertise and computing power. That Sabine herself can't see how such a system would work is merely a failure of imagination on her part. Not to mention making unreasonable assumptions about what is or isn't possible for an advanced civilization with effectively unlimited resources. That said, she is strictly correct when she says its not a scientific idea. By that its meant that its not something you can test by any method we are currently aware of. That doesn't mean that its either wrong or that it will never be scientific, just that its beyond science for the moment. That's why its philosophers (and engineers who fancy themselves philosophers) you mainly see talking about it, not scientists.

* To simulate a universe, you must know all the laws of that universe (or at least have theories you wish to test) to the smallest scale you wish to simulate. This means you must have a Grand Unified Theory fully functional, something that some of the brightest minds are not certain we can actually achieve. * To simulate consciousness to itself you must know all facets of what consciousness is and how it works. Something that every time we think we have it figured out we find something new that totally invalidates large sections of what we thought we knew. * To run these calculations you likely need computers that border on planetary construction sizes and harvest significant portions of stellar output as well as construction on the nano scale. To me, such capabilities indicates a level of advancement that it goes beyond the Clarkian concept of any civilisation advanced enough over another would seem like magic to functionally being gods. And if the simulation is so perfect that the simulated can't tell they are simulated does it matter? And given the concept that is used to make it oh,so likely of simulation in a simulation in a simulation, what kind of power must be ascribed to those at the top of the chain unless we assume that a simulation can be made such that it has more power to simulate within it than the simulation itself? I would say, however, that many theories needs must rely on some degree of "faith". Let us take evolution vs. Evolution. The idea that one creature over successive generations adapts to better fit an environment can be clearly seen and demonstrated. The process by which one species becomes another, speciation, is not well understood and some species that seem to be in the process are of great interest to help us explain it. The greater the leap the less we know. By the time we reach how a colony of single cells becomes a multi-cellular organism we have at best speculation and how a molecule that happily reproduces on its own in the right chemical soup developed all the machinery to build a cell around it seems in the realm of "Well, if it didn't we wouldn't be here to wonder about it.". So to confidently state that all life is the result of random chemical processes and changes demands a belief in something held to be true that cannot be demonstrated to the natural senses nor arrived at logically without already accepting the premise which is a definition of faith.

I'm not worried that someone might pull the plug, I'm terrified that it might be running on Windows.

Finally! I am so tired of the simulation argument being spammed everywhere and presented as if was an established fact. What angers me the most about it is that it's not only unscientific, but also completely and utterly useless, just like multiverse. On the other hand, it gives me hope: if smartest people can believe in such bullshit, maybe I am not so dumb after all.

Existance of peeing, pooping and farting is really say something about the civilization who made us. I wonder how did the conversation go.

Sadly, from 4:50 on, this video demonstrates a misunderstanding of how virtual clocks work and how virtualization in general works, even on our current “primitive” computer level. Sorry to say that. I don’t mean to say we live in a simulation, because I don’t think it matters and I don’t think we have a way to find out. I don’t believe in the simulation argument. I’m just explaining why one specific counter-argument presented here is wrong. The assumptions at 7:05 and afterwards are flat out wrong. Here’s an oversimplified summary of a way to avoid an “observable inconsistency” perceived by the simulation’s guest environment. - Stating the obvious: When you simulate a universe with civilizations, you observe, in real time, whatever they “observe”. - Each time they “observe” something you don’t like, you stop the simulation, fix your virtualized environment, rewind the simulation to a snapshot before their “observation”, restart the simulation. Rinse and repeat. - Understanding the step above goes back to my introductory gripe about (mis)understanding of virtual clocks. - On the next level, to make the simulation progress faster with fewer stops and rewinds, the simulation can be forked a massive number of times with slight variations introduced to each replica. The most “promising” branches (think of them as UNIX processes with fork() if it helps your intuition) are allowed to develop further whereas the ones that malfunction, “collapse” or perhaps expose virtualization to the guest environment are stopped (what an euphemism) to free up computing resources for the more promising ones. That’s all, that’s it. There goes the “observable inconsistency” argument. Notice that there’s nothing that would require this ”simulation development model” to cover the size spectrum in its entirety, all sizes at once. This is (again) because there is no need to avoid an “observable inconsistency” at all cost. When you spot an inconsistency - next branch, please / last snapshot, please! No big deal.

I think this (saying the simulation hypothesis is pseudoscience) is a classical example of mixing logic/philosophy (or other not empirical science) with empirical science. As far as I know, the simulation hypothesis is a form of the "brain in the vat", or originally "evil demon" argument from Decsartes. This argument is used to impress upon us that we cannot know anything for certain except the fact that a thing that is thinking these thoughts at this moment exists ("I" or "me"). Everything else is uncertain. This is not an empirical argument per se but much more a logical one. If you mix this epistemological argument with probability (which is a very contentious thing in itself as far as I am aware of) you get the simulation argument. The point of the argument is, I think, that we cannot possibly ever know whether we are living in a simulation or not (even if it should be "likely" that we do live in a simulation according to Bostroms argument). Concerning the physical arguments from the video I am not convinced: only 200, or say, 500 to be on the safe side, years ago nobody in their right mind thought we could split atoms, fly to the moon, have video-meetings or simulate the birth and life of galaxies (which we didn't know existed) on computers etc. etc. So of course we would today think that such simulations (as suggested by Bostrom) are not possible. Who knows what will be possible in 100 or 1000 years time? So the simulation hypothesis is/can be logically or philosophically quite scientific. It is just not an empirical statement, a statement about the real world ("we do live in a simulation"). But to call something not empirical is not the same as it being unscientific: nobody would say that mathematics is pseudoscience though all of it is grounded in unprovable and completely abstract axioms. It is very sad, I completely agree, that people who should know better, like Neil deGrasse Tyson do not and say stuff like that (i.e., derive empirical statements from non-empirical ones). This is quite scientifically illiterate imo. But I have heard "proven" or "bewiesen" in an empirical context so often from so many otherwise smart people, my ears have bled all out. One last thing: pseudoscience is very, very hard to define... This is called the problem of demarcation and it has not been solved, as far as I know. But it's fun to think about! :) Like the simulation hypothesis or the problems of the definition and meaning of probability and all that good stuff :D

“If you thought that science was certain - well that is just an error on your part.” Richard Feynman.

We live in a seemless, physical world, created by an allpowerful being, that lives in a paralell reality, that can act in our reality, but we can not see.

If we find a "bug" in the system, this could be an indication that, maybe, this hypothesis is not so absurd. Presumably, the coders could have made a coding error.

This is how science communications should be done. I love your channel.

Nerd scientist: simulation hypothesis Occam: hold my razor

After carefully contemplating this I have decided there will never be a computer capable of simulating Sabine.

Let’s expand what we mean by “simulation.” Because every time we dream, we enter into a type of simulated reality complete with unique story lines and characters. The simulation in this case is being run by our own minds. If we becomes lucid to the fact that we’re dreaming we can take control of the laws of nature within the simulation. Perhaps this reality is a simulation “of the mind” much like when we dream.

No need to simulate an entire universe - just simulate a single consciousness.

“I dare you” to god shutting off the simulation is priceless I subscribed immediately

The only question is why does it seem like that things are changing and we might discover new laws in the future which in turn allow the development of currently impossible technologies? Because that does seem to be a thing. There seems to be a feedback between human understanding and perception and reality, but maybe that's just an anthropocentric view. Anyone?

I think the connection to fundamentalism is actually at the core of this appealing bit of sci-fi which is really frankly nothing more than the latest reification of our concepts of how computation explains everything ( see Steven wolfram's work). A hidden being pulling the cosmic strings. And contrary to what people think, we have all had actually direct and real experience with these nearly omnipotent super beings - our preverbal experience as infants being taken care of by our parents. This has been called rather poignantly "the unrememberable and the unforgettable". These super beings, our parents at an early age for us, could make things magically all better and occasionally disastrously worse, very much like gods and demons operating outside of natural law because we had no sense at all at that time of how the physical world really worked or what caused what. Although we think of magical thinking as coming from some delusional source, it actually has this developmental basis. And we regress to that early developmental level in times of stress or when we are indulging our deepest wishes. There is also the reification of a single explanatory concept in this case computation. We are no closer to explaining consciousness in terms of anything that could be computed than we are at exceeding the speed of light in a spaceship. Both are simply science fiction fantasies at this point, and most of the work on computation completely ignores the basic principle in biology, namely that something is selected because it enhances survival. In other words Consciousness emerged because it helped to preserve homeostasis, and the evidence for this is simply overwhelming in that classic homeostatic imbalances have a priority and as it were basement level access to the Machinery of consciousness, and is consistent with the reticular and hypothalamic systems forming the neural foundations for an embodied self. Last but not least all, these concepts express a fundamental arrogance as opposed to a fundamental humility about nature. We have so much to learn, we are so ignorant, and there are so many Mysteries. Instead these Concepts make it sound as though we've got it all figured out. Even if the answer that it yields is a bit bizarre. In any case I agree with Sabine that these Notions are mostly untestable, although the available evidence argues against this is being anything more than a scientific cultural meme expressing our reification of computational concepts, and a regression to magical thinking . Judging from the posts however passionately defending this fundamentally magical idea, it does have some powerful hooks into our psychology